Hindawi Publishing CorporationJournal of Biomedicine and BiotechnologyVolume 2012, Article ID 159578, 6 pagesdoi:10.1155/2012/159578

Research Article

Proximity of Residence to Bodies of Water and Risk for West NileVirus Infection: A Case-Control Study in Houston, Texas

Melissa S. Nolan,1 Ana Zangeneh,1 Salma A. Khuwaja,2 Diana Martinez,3

Susan N. Rossmann,4 Victor Cardenas,1 and Kristy O. Murray1

1 Center for Infectious Diseases, School of Public Health, The University of Texas Health Science Center at Houston,Houston, TX 77030, USA

2 Office of Surveillance and Public Health Preparedness, Houston Department of Health and Human Services,Houston, TX 77054, USA

3 Office of Disease Control and Clinical Prevention, Harris County Public Health & Environmental Services,Houston, TX 77027, USA

4 Gulf Coast Regional Blood Center, Houston, TX 77054, USA

Correspondence should be addressed to Kristy O. Murray, kristy.o.murray@uth.tmc.edu

Received 1 August 2011; Revised 27 September 2011; Accepted 19 October 2011

Academic Editor: Roy A. Hall

Copyright © 2012 Melissa S. Nolan et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

West Nile virus (WNV), a mosquito-borne virus, has clinically affected hundreds of residents in the Houston metropolitan areasince its introduction in 2002. This study aimed to determine if living within close proximity to a water source increases one’sodds of infection with WNV. We identified 356 eligible WNV-positive cases and 356 controls using a population proportionateto size model with US Census Bureau data. We found that living near slow moving water sources was statistically associatedwith increased odds for human infection, while living near moderate moving water systems was associated with decreased oddsfor human infection. Living near bayous lined with vegetation as opposed to concrete also showed increased risk of infection.The habitats of slow moving and vegetation lined water sources appear to favor the mosquito-human transmission cycle. Thesemethods can be used by resource-limited health entities to identify high-risk areas for arboviral disease surveillance and efficientmosquito management initiatives.

1. Introduction

Houston, Texas, is a metropolis in the southeastern UnitedStates with around four million residents [1]. West Nilevirus (WNV) human cases were first reported locally in2002 [2] and have since become endemic with human casesreported annually [3]. WNV is an arboviral disease fromthe Flaviviridae family whose main transmission cycle occursbetween birds and mosquitoes; humans serve as an incidentalhost. In southeastern United States, Culex quinquefasciatusmosquitoes have been demonstrated as important vectors ofWNV disease transmission [2, 4].

In the United States, WNV transmission season tra-ditionally occurs from spring to fall, with a peak in latesummer [2]. In warm weather, mosquito larval developmentoccurs within days [5, 6] allowing for rapid reproduction

of new mosquito populations. Mosquito larval developmentoccurs in water bodies with each species having their ownpreferential type. Culex quinquefasciatus mosquitoes have adiverse larval habitat range, with high larval counts nearhuman habitation [7, 8]. Mosquito control efforts in Hous-ton, Texas, target residential areas where either mosquitopools or dead birds are positive for WNV disease. Targetedareas are identified through random mosquito trapping andreporting of dead birds by residents. The ecological dynamicbetween vector, reservoir, and human habitats is critical tounderstand when examining risk for human WNV infection.While this vector’s larval habitat preferences are known, nostudies to date have examined direct associations betweenlarval water habitats and WNV human disease transmission.This paper presents a novel method for examining diseaseclustering and its spatial association with water sources.

2 Journal of Biomedicine and Biotechnology

2. Methods

A case-control study design was used to determine theassociation between water sources and the risk of humaninfection with WNV.

2.1. Case Selection. Cases were defined as WNV-positivepatients identified through local surveillance performed bythe Houston Department of Health and Human Services(HDHHS), Harris County Public Health and EnvironmentalServices (HCPHES), or the Gulf Coast Regional BloodCenter (GCRBC). Local surveillance identified cases either bystate mandatory reporting laws or by national blood dona-tion testing guidelines that required laboratory confirmationof WNV human disease. Previous research has shown thatthe highest rates of WNV human seroprevalence were amongthose who reported a history of being outside during thehours of dusk and dawn [9]. These hours are concurrent withthe peak activity time of Culex quinquefasciatus mosquitoes.Since most people are at home during dusk and dawn, itwas resolved that cases are most likely exposed while athome. It was determined appropriate to use cases’ homeaddress at time of disease development as their location ofmosquito exposure. Cases’ home addresses were collectedvia case investigations performed by HDHHS, HCPHES, orGCRBC during 2002 and 2009. Exclusion criteria includedevidence of nonlocally acquired disease as documented in thecase investigation nonrecognition of address by MapMarkerUSA version 14 geocoding software, or home address fallingoutside the metropolitan’s geographic area as determined bythe geocoding software. After applying the exclusion criteria,we had 356 residential addresses from cases for final analysis.

2.2. Control Selection. Controls were defined as selectedblock centroids generated from the United States CensusBureau decennial data (http://www.census.gov/). Controlswere selected using two methods: a population proportionateto size sampling method which takes into account varyingpopulation densities within the metropolitan city and arandom sampling method. There were three selection framesthat were used to identify the final control. In descendingorder the frames were census tract level, block group, andfinally block. The population proportionate to size samplingmethods was used to select the initial frame: census tractlevel. It was understood that population distribution wasuniform throughout the census tracts selected; therefore,we used a random selection method for the two additionalframes: block group and block. Since the smallest definedcensus level is a block, the centroid of the block level wasused as a surrogate for control households. Based on samplesize calculations, a 1 : 1 case-control ratio was determinedappropriate to satisfy statistical significance using disciplinestandards; therefore, 356 control addresses were selected forfinal analysis.

2.3. Data Analysis. Spatial analysis of case and controlresidential distances’ to local water body sources was per-formed using MapInfo v9.5.1 software. Shapefiles of water

sources within the metropolitan’s geographic parameterswere provided in kind by Dr. Irina Cech, professor atthe University of Texas Health Science Center at Houston.The shapefiles were based on United States GeologicalSurvey water source definitions and data. Case and controlresidential coordinates were superimposed onto the watersource shapefile. Water source labels were used to identifythe particular water source, that is, Cedar Spring, Lou River,Brays Bayou, and so forth. The water source type was inferredfrom these labels. Using the software’s measurement tool,we measured the distance from each case/control point, tothe closest water source, excluding salt water sources sinceCulex quinquefasciatus mosquitoes do not utilize salt watersources as larval habitats [5]. For each case/control point werecorded the proximity to the closest water source, the typeof the particular water source, and the name of the partic-ular water source. We used STATA v11.0 (College Station,Texas) to run all statistical analyses. Chi-squared tables andlogistic regression were used to analyze the significance ofproximity to a water source between the two populations.Odds ratios, 95% confidence intervals (CIs), and P valueswere computed to analyze the significance of three factors:specified residential proximity to a water source; proximity toa particular water source type; proximity to a particular watersource. Attack rates (number of WNV human cases over totalnumber of households) were calculated for each census tractand mapped to spatially identify areas of high WNV humantransmission. A Getis Ord hot spot analysis was performedusing ESRI ArcGIS 10.0 to determine concentrations ofhigh and low human disease clustering. The GetisOrd (Gi)hot spot analysis identifies clusters of higher and lowermagnitude than would be randomly found and statisticaloutput is in the form of a Z score known as a GiZ score. Areasof high clustering were indicated by a GiZ score of 1.96 orgreater, and areas of low clustering were indicated by a GiZscore of −1.96 or less.

3. Results

On average, cases and controls resided the same proximityfrom water sources [x0 (controls) = 892 meters, x1 (cases) =931 meters]. Using linear regression, we found no statisticalassociation between residential proximity to water and oddsfor human WNV infection. However, when we binomially-coded at varying distances ranging from 50 to 750 meters, wefound a significant protective trend from distances rangingfrom 50 to 200 meters (Table 1). Living less than or equal to200 meters from a water source (x2 = 6.67, P < 0.01) wasfound to be protective from infection by a factor of 0.54.

Water source types were analyzed for association withodds for human WNV infection using odds ratios and chi-squared tests, as seen in Table 2. We examined the six mostcommon water source types. Two water source types werestatistically associated with odds of human infection. Livingnear a creek increased one’s odds of human infection by afactor of 1.37 (P = 0.09). Living near a spring decreased one’sodds of human infection by a factor of 0.55 (P = 0.06). Tofurther analyze these associations, we created two groupings

Journal of Biomedicine and Biotechnology 3

Table 1: Distance of case residence compared to US Census controlcentroids to water source in meters, evaluated by odds ratio (OR),95% confidence intervals (CI), and significance (P value).

Distance (m) OR 95% CI P value

50 0.10 (0.01, 0.42) <0.01

100 0.21 (0.07, 0.42) <0.01

150 0.35 (0.18, 0.66) <0.01

200 0.54 (0.32, 0.89) 0.01

250 0.70 (0.46, 1.05) 0.07

300 0.76 (0.52, 1.11) 0.14

350 0.78 (0.54, 1.12) 0.16

400 0.82 (0.58, 1.16) 0.24

450 0.82 (0.59, 1.14) 0.22

500 0.85 (0.62, 1.17) 0.31

550 0.87 (0.63, 1.19) 0.35

600 0.78 (0.57, 1.07) 0.11

650 0.84 (0.62, 1.15) 0.26

700 0.89 (0.66, 1.21) 0.45

750 0.92 (0.68, 1.25) 0.60

based on slow moving and moderate moving water sourcetypes. A grouping of slow moving water bodies (creeks andgullies) was found to increase one’s odds of human infectionby a factor of 1.45 (P = 0.03). A grouping of narrowmoderate moving water bodies (streams and rivers) wasfound to be protective against human infection by a factorof 0.50 (P = 0.02).

Particular water sources were evaluated for associationwith odds for human WNV infection by odds ratios and chi-squared tests, as seen in Table 3. The eleven most commonspecific water sources were analyzed. Two water body sourceswere significantly associated with increased odds for humaninfection. Living close to White Oak Bayou (P = 0.01)increased one’s odds of human infection by a factor of 2.25.Additionally, living near Cypress Creek (P = 0.02) wasalso associated with increased odds of human infection bya factor of 2.54. Since Cypress Creek has several tributaries,an additional category was made that included all feeders forCypress Creek. This group had the strongest significance ofall water bodies (P < 0.01) with increased odds of humaninfection by a factor of 1.93. We also found that living closeto Buffalo Bayou had increased odds of human infection bya factor of 1.59, which neared significance (P = 0.07).

Spatial distribution of WNV attack rates per 10,000population by census tract illustrates that the highest riskarea of transmission is in Northwest Houston as seenin Figure 1. Hot spot analysis confirmed that there weresignificant clusters of cases in Houston as seen in Figure 2.The areas of highest valued clusters were along the Northwestcorner of Harris County, which overlaps Cypress Creek andits feeders. Figure 3 demonstrates the spatial relevance of theHouston area inlaid within Harris County, in relation to thestate of Texas, and the United States of America.

Table 2: Proximity of residence to water source types in cases versuscontrols, evaluated by odds ratio (OR), 95% confidence intervals(CI), and significance (P value).

Water source type∗ OR 95% CI P value

Bayou 1.15 (0.84, 1.56) 0.36

Creek 1.37 (0.93, 2.02) 0.09

Ditch 0.49 (0.13, 1.60) 0.19

Gully 1.50 (0.73, 3.16) 0.23

Lake 1.50 (0.73, 3.16) 0.23

Stream 0.55 (0.27, 1.08) 0.06

Creek and gully 1.45 (1.02, 2.07) 0.03

Stream and river 0.50 (0.25, 0.95) 0.02∗

As defined by the United States Geological Survey.

Table 3: Proximity of residence to particular water sources in casesversus controls, evaluated by odds ratio (OR), 95% confidenceintervals (CI), and significance (P value).

Particular water source OR 95% CI P value

Bering Ditch 0.66 (0.14, 2.82) 0.52

Berry Bayou 1.00 (0.26, 3.78) 1.00

Brays Bayou 0.73 (0.43, 1.23) 0.21

Buffalo Bayou 1.59 (0.93, 2.75) 0.07

Cypress Creek 2.54 (1.10, 6.35) 0.02

Cypress Creek and tributaries 1.93 (1.14, 3.33) 0.01

Greens Bayou 0.66 (0.26, 1.59) 0.31

Halls Bayou 1.00 (0.40, 2.47) 1.00

Hunting Bayou 1.89 (0.69, 5.66) 0.17

Little White Oak Bayou 1.81 (0.74, 4.72) 0.15

Sims Bayou 0.57 (0.19, 1.61) 0.25

White Oak Bayou 2.25 (1.15, 4.55) 0.01

4. Discussion

This is the first known case-control study to perform aspatial analysis of human WNV infection risk with regard toproximity of residences to water sources serving as surrogatesfor potential aquatic larval habitats. Overall, we found nodirect association between proximity of residences to watersources and odds of WNV human infection in Houston,Texas. However, we found a significant trend of decreasedrisk of infection among people living within 200 metersof a water source. It is conjectured that areas closest towater sources are the primary target of mosquito controlprograms, therefore decreasing the risk of transmissionat closer distances. We did find a pattern of increasingodds ratios as distance increased by 50-meter intervals,suggesting that mosquitoes in Houston have an expansiveflight range that is important in the ecology of diseasetransmission. Culex quinquefasciatus mosquitoes are knownto have an expansive flight range with recapture documentedup to 1000 meters outside of their release site [10]. Onespeculation could be that the use of adulticides along waterbodies could temporarily suspend adult mosquito activityallowing for higher mosquito activity occurring at greaterdistances. Although adulticides are the primary mosquito

4 Journal of Biomedicine and Biotechnology

Attack rate by census tract

5.3 to 14.12.4 to 5.30.5 to 2.40 to 0.5

N

Figure 1: Attack rate: number of reported West Nile virus cases per 10,000 population using 2000 US census tract data in the Houstonmetropolitan area, Texas.

GiZ score

<−2.58 std. dev.−2.58–−1.96 std. dev.−1.96–−1.65 std. dev.−1.65–1.65 std. dev.1.65–1.96 std. dev.1.96–2.58 std. dev.>2.58 std. dev.

Figure 2: Hot spot cluster analysis of West Nile virus cases in the Houston metropolitan area, Texas.

control method used in this area, it is known that the useof adulticides is random and not associated with specificwater bodies. Another speculation is that alternate breedingsites, specifically storm sewers, also play a role in diseasetransmission. In Houston, Culex quinquefasciatus are thedominate mosquito species collected from storm sewers, andstorm sewers have been demonstrated as a preferential sitefor breeding, larval development, and daytime resting [11].Unfortunately, we did not have access to sewer blueprints ofthe metropolitan area to further investigate this theory.

When analyzing residential proximity to water sourcetypes, we did find a strongly significant association for

risk of human infection among residences near creeksand gullies, specifically Cypress Creek. It is believed thatthe slower movement of water and dense vegetation ispreferential for the local transmitting Culex vector species.Due to low numbers of cases per creek, no additionalspecific creek sources were included in the final analysis.Cypress Creek is a large water source that flows throughoutthe northwest corner of the metropolitan Houston area.Figure 1 demonstrates that attack rates of human infectionare strongest in the area where Cypress Creek flows. Thisfinding is further substantiated by Figure 2, which showsthe highest clusters of human WNV cases are in the area

Journal of Biomedicine and Biotechnology 5

Harriscountry

State of Texas

Figure 3: Geographic location of metropolitan Houston area inlaid within Harris County in relation to the State of Texas and the UnitedStates of America.

where Cypress Creek flows. We feel the true association ofinfection is with the particular water source Cypress Creek.Additional studies should perform mosquito pool testingaround Cypress Creek and additional creeks and gulliesthroughout the metropolitan area to examine WNV fieldinfection rates of mosquitoes in efforts to further validate ourfindings.

When analyzing residential proximity to water sourcetypes, we did find a strong protective association of res-idences closest to streams; however, no particular streamwater sources were identified as being associated withinfection. To further investigate these findings, we created agrouping of moderate moving water sources which includedstreams and rivers. This grouping had the strongest signifi-cance of protection from human WNV infection. Addition-ally, no particular river water sources were identified as beingassociated with infection. These findings are evidence thatresidences in closest proximity to moderate moving watersources are significantly protected against WNV humaninfection.

Houston is prone to flooding, and as part of the floodmitigation program, the city has an extensive network ofbayous, which are man-made canals [2]. The surroundinghabitats of bayous in Houston are varied with some beingcast with concrete walls and others edged with grass, shrubs,and other vegetation. Overall, we did not find an associationbetween the living near bayous and increased odds ofinfection. However, we did find that White Oak Bayou andBuffalo Bayou were significantly associated with increasedodds of infection. These specific bayous are lined withextensive vegetation preferential to mosquito habitats. Thisis in sharp contradiction to the bayous lined with concrete,such as Brays Bayou, where the data suggested decreasedodds of infection. We cogitate that the type of bayou liningand habitat dictates WNV transmission. Future researchshould incorporate bayou linings and their individual risk forlocal human habitants.

There are a few limitations of this study that are worthnoting. One limitation was the potential for selection biasdue to the inability to verify disease status of controlsby serum antibody testing. Since WNV is a mandatory-reportable disease in the state of Texas, anyone who testedpositive should have been reported to the local health depart-ment. The risk of misclassification of controls is possible ifa resident at the address never developed symptoms or hadmild disease that went undiagnosed as WNV. However, thisrisk is presumed minimal since current estimates of sero-prevalence in Houston are relatively low [12]. Due to finan-cial constraints, we were unable to obtain a serum samplefrom controls to verify disease status. Lastly, we were unableto test for potential confounders related to human-mosquitotransmission, such as socioeconomic status, gender, rainfall,or other seasonal environmental factors. Complete recordsfor these potential confounders were unavailable. Despite theinability to control for these potential confounds, we believethe results are sound considering people do not choose theirresidence location based on human-mosquito transmissionhotspots.

The main strength of the study is the ability to determinehigh risk areas of WNV transmission around the Houstonmetropolitan area using minimal resources. The methodswe used are simple to perform and could be of benefitto health authorities in other jurisdictions to identify areaswith increased risk for WNV transmission. In resource-scarce public health departments, this inexpensive methodcould greatly increase the effectiveness of mosquito controlprograms. Our case-control selection methods would besimple to replicate. Since WNV is a reportable diseasenationally, case investigations are performed for all patientsthat test positive. From these case investigations, healthdepartments should have the addresses of the cases in theirjurisdiction. Control selection would be easy to execute ascensus data is readily available from the US Census Bureauwebsite that is updated both annually and decennially.

6 Journal of Biomedicine and Biotechnology

In conclusion, we found that living near slow movingwater bodies, such as creeks and gullies, or bayous withheavy vegetation increased one’s odds of infection withWNV. Most importantly, we identified Cypress Creek asan area of high WNV human infection that should betargeted by future mosquito control efforts. With the recentliterature suggestive of increased ranges of arboviral vectorsand areas of transmission, this method of spatial analysiscould benefit other health authorities in areas experiencingactive WNV transmission who need predictive models ofexposure risk for targeted education and control efforts fordisease prevention.

Conflict of Interests

The authors have no conflicts of interests to report.

Acknowledgments

The authors thank the cohort participants for their contribu-tion to this study, as well as Dr. Keith Bureau, Dr. Irina Cech,Dr. Adebowale Awosika-Olumo, and staff from the City ofHouston Department of Health and Human Services, HarrisCounty Public Health and Environmental Services, andGulf Coast Regional Blood Center for their assistance. Thiswork was funded in part through a grant from the UnitedStates Department of Defense, Telemedicine and AdvancedTechnology Research Center (TexSHIELD W81XWH-07-2-0031), and the Gillson-Longenbaugh Foundation. This studywas approved by the University of Texas Health ScienceCenter at Houston Committee for the Protection of HumanSubjects (HSC-SPH-03-039) in collaboration with the otherinstitutions involved in this project.

References

[1] U S census bureau, 2010, http://www.census.gov/.[2] K. M. Lillibridge, R. Parsons, Y. Randle et al., “The 2002

introduction of West Nile virus into Harris County, Texas, anarea historically endemic for St. Louis encephalitis,” AmericanJournal of Tropical Medicine and Hygiene, vol. 70, no. 6, pp.676–681, 2004.

[3] USGS disease maps, 2010, http://www.diseasemaps.usgs.gov/.[4] R. B. Tesh, R. Parsons, M. Siirin et al., “Year-round West

Nile virus activity, Gulf Coast Region, Texas and Louisiana,”Emerging Infectious Diseases, vol. 10, no. 9, pp. 1649–1652,2004.

[5] R. Matheson, Handbookof the Mosquitoes of North America,Hafner Publishing Company, New York, NY, USA, 2ndedition, 1966.

[6] J. B. Henn, M. E. Metzger, J. A. Kwan et al., “Development timeof Culex mosquitoes in stormwater management structuresin California,” Journal of the American Mosquito ControlAssociation, vol. 24, no. 1, pp. 90–97, 2008.

[7] E. J. Muturi, J. I. Shililu, B. G. Jacob et al., “Diversity ofriceland mosquitoes and factors affecting their occurrenceand distribution in Mwea, Kenya,” Journal of the AmericanMosquito Control Association, vol. 24, no. 3, pp. 349–358, 2008.

[8] C. M. Barker, B. G. Bolling, C. G. Moore, and L. Eisen,“Relationship between distance from major larval habitats and

abundance of adult mosquitoes in semiarid plains landscapesinColorado,” Journal of Medical Entomology, vol. 46, no. 6, pp.1290–1298, 2009.

[9] F. Mostashari, M. L. Bunning, P. T. Kitsutani et al., “EpidemicWest Nile encephalitis, New York, 1999: results of a household-based seroepidemiological survey,” The Lancet, vol. 358, no.9278, pp. 261–264, 2001.

[10] W. K. Reisen, M. M. Milby, and R. P. Meyer, “Populationdynamics of adult Culex mosquitoes (diptera: culicidae) alongthe Kern River, Kern County, California, in 1990,” Journal ofMedical Entomology, vol. 29, no. 3, pp. 531–543, 1992.

[11] R. Parsons, “Mosquito control-Texas style,” Wing Beats, vol.14, pp. 4–38, 2003.

[12] K. O. Murray, S. Baraniuk, M. Resnick et al., “Risk factorsfor encephalitis and death from West Nile virus infection,”Epidemiology and Infection, vol. 134, no. 6, pp. 1325–1332,2006.